Download www.drmichaellevin.org

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

List of types of proteins wikipedia , lookup

Node of Ranvier wikipedia , lookup

Signal transduction wikipedia , lookup

Cellular differentiation wikipedia , lookup

JADE1 wikipedia , lookup

Hedgehog signaling pathway wikipedia , lookup

VLDL receptor wikipedia , lookup

Sonic hedgehog wikipedia , lookup

Transcript
DEVELOPMENTAL BIOLOGY 189, 57–67 (1997)
ARTICLE NO. DB978662
Left/Right Patterning Signals and the Independent
Regulation of Different Aspects of Situs
in the Chick Embryo
Michael Levin,*,1 Sylvia Pagan,* Drucilla J. Roberts,*,†
Jonathan Cooke,‡ Michael R. Kuehn,§
and Clifford J. Tabin*,2
*Department of Genetics, Harvard Medical School, 200 Longwood Avenue, Boston,
Massachusetts 02115; †Department of Pathology, Massachusetts General Hospital,
Boston, Massachusetts 02114, and Division of Women’s and Perinatal Pathology,
Brigham and Women’s Hospital, Boston, Massachusetts 02115; ‡Laboratory
of Developmental Neurobiology, National Institute for Medical Research,
The Ridgeway, Mill Hill, London NW7 1AA, United Kingdom; and
§Experimental Immunology Branch, National Cancer Institute,
National Institutes of Health, Bethesda, Maryland 20892
Recently, a pathway of genes which are part of a cascade regulating the side on which the heart forms during chick
development was characterized (M. Levin et al., 1995, Cell 82, 1–20). Here we extend these previous studies, showing that
manipulation of at least one member of the cascade, Sonic hedgehog (Shh), can affect the situs of embryonic rotation and
of the gut, in addition to the heart. Bilateral expression of Shh, which is normally found exclusively on the left, does not
result in left isomerism (a bilaterally symmetrical embryo having two left sides) nor in a complete situs inversus phenotype.
Instead, misexpression of Shh on the right side of the node, which in turn leads to bilateral nodal expression, produces a
heterotaxia-like condition, where different aspects of laterality are determined independently. Heart situs has previously
been shown to be altered by ectopic Shh and activin. However, the most downstream gene identified in the LR pathway,
nodal, had not been functionally linked to heart laterality. We show that ectopic (right-sided) nodal expression is able to
affect heart situs, suggesting that the randomization of heart laterality observed in Shh and activin misexpression experiments is a result of changes in nodal expression and that nodal is likely to regulate heart situs endogenously. The first
defined asymmetric signal in the left–right patterning pathway is Shh, which is initially expressed throughout Hensen’s
node but becomes restricted to the left side at stage 4/. It has been hypothesized that the restriction of Shh expression
may be due to repression by an upstream activin-like factor. The involvement of such an activin-like factor on the right
side of Hensen’s node was suggested because ectopic activin protein is able to repress Shh on the left side of the node, as
well as to induce ectopic expression of a normally right-sided marker, the activin receptor cAct-RIIa. Here we provide further
evidence in favor of this model. We find that a member of this family, Activin bB, is indeed expressed asymmetrically, only
on the right side of Hensen’s node, at the correct time for it to be the endogenous asymmetric activin signal. Furthermore,
we show that application of follistatin-loaded beads eliminates the asymmetry in Shh expression, consistent with an
inhibition of an endogenous member of the activin–BMP superfamily. This combined with the previous data on exogenous
activin supports the model that Activin bB functions in the chick embryo to initiate Shh asymmetry. While these data
extend our understanding of the early signals which establish left–right asymmetry, they leave unanswered the interesting
question of how the bilateral symmetry of the embryo is initially broken to define a consistent left–right axis. Analysis
of spontaneous chick twins suggests that, whatever the molecular mechanism, left–right patterning is unlikely to be due
to a blastodermal prepattern but rather is initiated in a streak-autonomous manner. q 1997 Academic Press
1
Current affiliation: Department of Cell Biology, Harvard Medical
School, Boston, MA 02115.
2
To whom correspondence should be addressed. Fax: (617) 4327595. E-mail: [email protected].
0012-1606/97 $25.00
Copyright q 1997 by Academic Press
All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$401
57
08-12-97 06:09:05
dbal
58
Levin et al.
INTRODUCTION
During development, embryos acquire complex patterns
along each of the three axes. The resulting morphologies
can display a variety of symmetry types, including spherical
(as in volvox), radial (as in starfish), chiral (as in snails),
bilateral (as in Drosophila), and pseudobilateral (as in man).
Most vertebrates have a generally bilaterally symmetrical
body plan, but this symmetry is broken into pseudosymmetry by the consistently asymmetric placement of various
internal organs such as the heart, liver, spleen, and gut, and
an asymmetric development of certain paired organs such
as brain hemispheres or lungs. While the specification of
anterior–posterior (AP) and dorsal–ventral (DV) axial asymmetries has been studied in some detail (reviewed in Hunt
and Krumlauf, 1992), far less is known about the molecular
mechanisms underlying left–right (LR) asymmetry.
The LR axis is probably specified after the anterior–posterior and dorsal–ventral axes and is determined with respect
to them (McCain and McClay, 1994; Danos and Yost, 1995).
Until recently, almost all available information on LR asymmetry centered around four lines of inquiry: a phenomenological literature describing various asymmetries (Neville,
1976, presents an extensive and fascinating survey), the genetics of chirality in snails (an unidentified cytoplasmic factor determines dextrality; Freeman and Lundelius, 1982),
several drugs which cause alterations in LR patterning (for
example, an adrenergic pathway is implicated by Fujinaga
and Baden, 1991a), and mammalian mutants which have
phenotypes associated with LR asymmetry, such as randomization or total reversal of internal situs (Brueckner et al.,
1989; Yokoyama et al., 1993). Selection for LR asymmetries
in Drosophila, in hopes of generating a genetically tractable
mutant, have been unsuccessful (e.g., Tuinstra et al., 1990).
The discovery of signaling molecules asymmetrically expressed prior to overt morphological LR asymmetry in vertebrates (Levin et al., 1995) has opened the way to understanding how the LR axis is regulated. In the chick embryo
Sonic hedgehog (Shh) is initially expressed throughout
Hensen’s node; however, it is subsequently expressed exclusively on the left, perhaps due to the influence of an activinlike activity on the right side of the node. Subsequently, the
resulting asymmetric expression of Shh induces the chick
homologue of the mouse gene nodal (a member of the TGFb family, previously called cNR-1), which spreads throughout the lateral plate mesoderm on the left side. Experimental manipulation of these asymmetrically localized signals
verified that they form a linear pathway (Levin et al., 1995).
Implanting a source of activin on the left side of the node
can repress Shh on the left; together with the normal absence of Shh in the right, this results in a lack of nodal
expression. Moreover, implanting a source of Shh on the
right side, where it is normally repressed, results in bilateral
expression of nodal. Either manipulation leads to randomization of heart situs, strongly suggesting that this molecular cascade is involved in the regulation of morphological
asymmetry.
Several important aspects of this pathway remained un-
clear, however. For example, while heart situs is clearly one
endpoint of this pathway, it is unknown whether it is a
heart-specific cascade or whether other aspects of laterality
(such as situs of the gut) utilize the LR information inherent
in the cascade. If this is the case, it becomes interesting to
ask whether perturbation of this pathway results in: (a) all
of the organs making coordinated, albeit randomized, situs
choice (situs inversus; Hummel and Chapman, 1959), (b)
the organs making independent laterality decisions in response to the signals (heterotaxia), or (c) an embryo formed
symmetrically with respect to the LR axis (isomerism). Situs inversus, heterotaxia, and isomerism all occur at a significant incidence in many vertebrates, including man
(Winer-Muram, 1995). Each of these conditions, in theory,
could result from symmetrical signaling. For example, leftisomerism (exemplified by polysplenia syndrome; Ivemark,
1955) might be predicted to result from double-sided Shh
expression, since Shh is a left-specifying factor. In this
study, we examine this question and address other upstream
and downstream steps in the LR-determining pathway.
MATERIALS AND METHODS
In Situ Hybridization
After being fixed in 4% paraformaldehyde overnight, chick embryos were processed for whole-mount in situ hybridization as described in Levin et al. (1995). The clones used in the Shh and nodal
in situ hybridizations are as described in Levin et al. (1995). The
nodal gene was previously called chick nodal-related-1 (cNR-1).
The nomenclature was changed to match that of the mouse gene
in recognition of the fact that the asymmetric expression of this
gene is shared among frogs, chicks, and mice (Levin et al., 1995;
Lowe et al., 1996) and based on parallels in function described in
this report. The DIG antisense probe for Activin bB (a gift from K.
Patel) covers a 341-bp fragment of the Activin bB clone. Embryo
staging was according to Hamburger and Hamilton (1951).
Nodal and Shh Misexpression
All experimental manipulations were performed on standard
pathogen-free white Leghorn chick embryos obtained from
SPAFAS (Norwich, CT). At the time these experiments were conducted our clone of chick nodal did not include the entire Nterminal pro portion of the protein which is removed in processing.
To obtain a construct that would encode a protein which would
be processed correctly and secrete a normal, mature nodal protein,
we fused the BMP-4 pro region (including the cleavage cite) to
the cNR-1 mature region. For misexpression, this construct was
inserted into the RCAS-BP(A) vector, which encodes a replicationcompetent retrovirus (Hughes, 1987). Chick embryonic fibroblast
(CEF) cells were infected with this construct (as in Riddle et al.,
1993) and pelleted. Pellets were implanted between the epiblast
and the hypoblast on the right side of stage 5–6 embryos in New
culture (New, 1955). In situ hybridization of tissue (including the
grafted pellet) to a cNR-1 riboprobe was used to ensure that the
cells produce cNR-1 mRNA (data not shown). Shh was misexpressed by implanting protein-soaked beads (270 mg/ml of N-terminus of human Shh protein produced in bacteria) on the right side
of stage 4–5 embryos in ovo.
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$401
08-12-97 06:09:05
dbal
59
Left/Right Patterning Signals
Extirpation of Presumptive Heart Region
The region of the 30-hr chick embryo fated to give rise to the
heart has been well mapped (Stalsberg and DeHaan, 1969). By this
time, these cells express several cardiac-specific markers, including
Nkx-2.5, but they have not yet started to form a morphological heart
tube. Embryos with 6–7 segmented somites (ca. 30 hr of incubation)
were explanted into New culture (New, 1955), but with the stretched
vitelline membrane left flattened, underlaid by only a small amount
of albumen medium and overlaid by a 2-mm layer of 1:1 Hanks
BSS:Liebovitz air-buffered TCM mixture. Under this condition,
groups of embryos could be assembled and age matched for experimental vs sham versions of the operation, which were carried out
with tungsten needles. Splanchic mesoderm and overlying endoderm/hypoblast of the precardiac regions were bilaterally excised up
to their midline junction at the developing gut pocked. Sham-operated controls received transverse cuts across these regions anteriorly
and posteriorly, but without removal of tissue. The space above
the embryo within the ring was then drained and that beneath the
vitelline membrane further filled with albumen to give convexity
of the cultured blastoderm. The embryos were then allowed to grow
in culture to the 18- to 20-somite stage, 20–24 hr later.
Follistatin Application
Heparin acrylic beads were soaked in follistatin protein (obtained
from the National Hormone and Pituitary Program) at approximately 0.05 mg/ml in water or PBS. Control beads were soaked in
water or PBS only. Beads were implanted between the epiblast and
the hypoblast of stage 3/ embryos in New culture (New, 1955) and
processed for in situ hybridization at appropriate later stages.
Analysis of Twins
Twins occur spontaneously in chick eggs at a rate of approximately 1–2%. Of such twins, approximately 5–10% are in the 1807
head-to-head orientation required for this experiment.
RESULTS
Shh Affects Situs of Multiple Organ Systems
It has been previously proposed (Waddington, 1937) that
heart looping mechanically sets the situs of other morphological aspects of laterality. To test this possibility we surgically removed the prospective heart region from six- to
seven-somite chick embryos and scored embryonic rotation. The stage when the heart tissue was removed was
prior to the formation of the heart tube, let alone the looping
of the tube which could exert physical forces on other organ
primordia. Inspection at stage 13–14, when embryonic rotation was assayed, verified that the heart was completely
removed by the surgical procedure. The results are summarized in Table 1. It is seen that the incidence of correct
embryonic rotation in embryos receiving a sham operation,
88% (Fig. 1A), is not significantly altered (x2 Å 0.126, P ú
0.5) in embryos having no heart whatsoever (76% correct
embryonic rotation, Fig. 1B). While this experiment does
not eliminate the possibility that cardiac cells signal other
organs to instruct laterality, it does argue strongly against
the possibility that the mechanical stress imparted by the
bending heart tube has this effect, as following our surgical
manipulations, no heart tube ever forms.
If the situs of organs other than the heart is not determined mechanically, then it could be set by a response to
signaling molecules. The asymmetric signals described in
early chick development have previously been shown to
affect heart situs (Levin et al., 1995). In principle, this signaling cascade could lie after the regulatory branch point specifically controlling heart morphogenesis or it could lie upstream, playing a more fundamental role in establishing LR
asymmetry of the body plan. The original studies of this
signaling pathway were carried out in vitro using the technique of New culture. This procedure allowed a large number of embryos to be surgically manipulated, since in ovo
surgery is much more difficult. However, using New culture, embryos do not survive long enough to assay the morphology of organs other than the heart.
To determine whether the activin–Shh–nodal pathway
is heart-specific or a general LR determination system, we
implanted beads soaked in Shh protein on the right side of
Hensen’s node (opposite its normal expression) in ovo. We
find that the in vivo procedure results in a much higher
mortality rate than in vitro. Moreover, a lower percentage
of surviving embryos show alterations in heart laterality
than in vitro (12.1% instead of 50%; Table 2). Both the
mortality and the decreased incidence of effect on heart
morphogenesis among survivors are likely to be due to the
deleterious long-term effects of successfully implanted Shh
beads over the extended time of incubation required to
allow other organs to develop.
We assayed surviving embryos for heart situs and for situs
of the stomach as a representative second asymmetric organ. In addition, we examined the direction of embryonic
rotation within the egg, a property which has previously
been suggested to be mechanically linked to the bending of
the heart tube (Waddington, 1937). For the purposes of this
study, inverted heart situs is defined as the heart tube’s
bending to the left, inverted stomach situs as the stomach
on the right, and inverted body rotation as the embryo turning to the left. Embryos were scored as inverted only when
unambiguous, not including, for example, embryos which
failed to rotate in either direction. This conservative approach results in underestimating rather than overestimating the effects of Shh on laterality decisions. In addition,
the only embryos scored were those which survived long
enough to assay other organs in addition to the heart.
Seventy-four experimental animals which survived the
procedure were examined 4 to 6 days after implantation.
We find that all three scored aspects of laterality (heart situs,
gut situs, and direction of embryonic turning) are affected
by the presence of ectopic Shh (Table 2). Approximately
10% of the treated embryos showed each individual type of
situs alteration. In a similar number of surviving embryos
implanted with control beads we observed only a 1.5% incidence of reversal in the direction of body rotation (1 embryo)
and no examples of reversal in heart or stomach situs. Those
controls were consistent with the low rates of spontaneous
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$402
08-12-97 06:09:05
dbal
60
Levin et al.
TABLE 1
Group
wt embryonic
rotation
Reversed embryonic
rotation
Ambiguous or absent
embryonic rotation
N
Statistics
Sham operation
Heart excision
88%
76%
6%
5%
6%
19%
15
21
x2 Å 0.126, P ú 0.5
Note. Heart territories from stage 8/ to 90 chick embryos in New culture were surgically removed (heart excision group) or transected
without removal of tissue (control group). Embryos were allowed to develop and the situs of the embryonic rotation was scored.
reversal in body rotation (about 1%) and reversal of internal
organ situs (less than 0.1%) we have observed in our laboratory. These results demonstrate that, in addition to heart
situs (Levin et al., 1995), bilateral exposure to Shh signaling
has a significant effect on rotation of the embryo (x2 Å 73,
P õ 0.005) and stomach situs (x2 Å 3.8, P õ 0.05). Thus
the asymmetric cascade lies upstream of the branch point
leading to laterality decisions for each individual organ.
Interestingly, the situs of the heart and stomach and direction of rotation appear to be influenced independently
by Shh signaling, resulting in a heterotaxic phenotype. We
found examples of treated embryos where each was the only
aspect of laterality which was reversed. Moreover, only 2
of the 16 embryos showing laterality defects were concordant for all three aspects being reversed (true situs inversus).
This is particularly surprising in the case of the direction
of heart looping and embryonic rotation as these have been
previously reported to be linked. However, in response to
Shh application we observed 5 cases where both the heart
FIG. 1. The Shh pathway independently controls situs of organs
other than heart. The prospective heart region was removed from
embryos at stage 6–7. Embryos were allowed to develop until embryonic rotation took place. Embryos receiving a sham operation
(A) rotate to the correct side in 88% of the cases; embryos whose
heart region has been extirpated (B) likewise rotate correctly in
76% of the cases. Dark gray arrowheads show torsion, light gray
arrowhead shows heart, and white arrowhead shows lack of heart.
and the embryonic rotation were reversed, 4 cases where
the heart but not embryonic rotation was reversed, and 4
cases where rotation but not heart looping was altered.
Thus, in our experiments, these two processes were affected
independently.
Nodal Influences Heart Morphogenesis
The results indicate that Shh signaling can influence the
situs of several asymmetric organs. This creates a paradox
however, since Shh is only asymmetrically expressed in a
very limited spatial and temporal domain in Hensen’s node
and is no longer asymmetrically expressed at the time when
the organs are formed. This suggests that the asymmetric
signaling by Shh must be mediated by a secondary signal.
Nodal is an excellent candidate for such a secondary signal.
Its asymmetric expression is induced by Shh (Levin et al.,
1995), correlating with the inductive effects of Shh on organ
laterality. Alteration in nodal expression has similarly been
correlated with changes in organ situs in the murine iv
(Lowe et al., 1996) and inv mutants (Collignon et al., 1996).
Moreover, the expression domain of nodal is quite broad,
initiating (in the chick) in the anterior lateral plate mesoderm and subsequently spreading posteriorly while retracting rostrally (Levin et al., 1995). At stage 8, it is expressed in a domain directly adjacent to cells expressing
Nkx-2.5, a marker of cardiac progenitor cells (Schultheiss et
al., 1995), consistent with nodal’s providing an asymmetric
signal to lateralize the heart primordia.
To test directly whether nodal is indeed capable of influencing heart situs, pellets of CEFs infected with a retroviral
vector expressing nodal were implanted into the right side
of embryos at stage 6–7, the stage at which endogenous
nodal is induced by Shh in the left side. To obtain higher
frequency of survival, the experiments were carried out in
vitro, in New culture (New, 1955). The results are summarized in Table 3. Under these conditions, in embryos receiving no implant or an implant of cells infected with a nonspecific control virus (alkaline phosphatase), 81% of the developing heart tubes bend to the right, as normal (Fig. 2A),
while 19% were either inverted (Fig. 2B) or bilaterally symmetric (Fig. 2C), as previously seen under New culture conditions. There was no significant change in the percentage
of hearts bending to the right (82%) when nodal-expressing
cells were implanted in the left side, where nodal is normally expressed; however, when nodal was misexpressed
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$403
08-12-97 06:09:05
dbal
61
Left/Right Patterning Signals
TABLE 2
Control beads
Heart situs
97% wt
0% reversed
2% abnormal
100% wt
0% reversed
97% wt
1.5% reversed
1.5% unturned
65
Stomach situs
Embryonic rotation
No. of embryos examined
Shh beads
87.8% wt
12.1% reversed
0% abnormal
90.5% wt
9.5% reversed
81.2% wt
13% reversed
5.8% unturned
74
Statistics
2
x Å 150
P Å 0.005
x2 Å 3.8
P Å 0.05
x2 Å 73
P Å 0.005
Total x2 Å 6.6, P õ 0.01
Note. Beads loaded with Shh protein by equilibrating for 24–48 hr were implanted into the right side of Hensen’s node of stage 4
embryos in ovo. Embryos were allowed to develop and the situs of the heart, gut, and embryonic rotation was scored.
on the opposite side, approximately half that number bent
to the right (38%), with a corresponding increase in both
inverted and bilaterally symmetric hearts. This effect is significant to P Å 0.005 (x2 Å 23.9). It should be noted that
approximately twice as many affected hearts were bilaterally symmetric, looping in both directions (right-isomerism,
43%) as were inverted (19%). Since ectopic nodal expression
can alter heart situs, these data implicate nodal as part of
the functional cascade determining cardiac laterality. The
ability of nodal to affect LR specification has been independently demonstrated in Xenopus (Sampath et al., 1997).
Signaling Upstream of Shh in the LR Asymmetric
Cascade
Identifying a series of signals directing the asymmetric
morphogenesis of the visceral organs is important, in part,
because it provides the opportunity to work backward toward addressing the origin of LR asymmetry during embryogenesis. The first described signal in the asymmetric
cascade is Shh, which is uniformly expressed at Hensen’s
node at stage 40, but then is asymmetrically repressed on
the right side at stage 4/. An upstream activin-like signal
mediating this repression was suggested by the finding that
an activin-inducible marker (the activin receptor cAct-RIIa)
is expressed on the right side of Hensen’s node concomitant
with Shh repression (Levin et al., 1995). Consistent with
an activin-like activity playing such a role, ectopic activin
protein applied to the left side of the node can repress Shh
and induce cAct-RIIa.
The model that an activin-like protein is involved in LR
determination thus depended heavily on the effects of
applying ectopic activin. If an endogenous activin-like signal is indeed critical for establishing the later asymmetric
expression of Shh and nodal, then interfering with such
signals at stage 3–4 should alter Shh and nodal expression.
To test this, we implanted beads loaded with follistatin, an
antagonist of signaling by activin and related molecules,
including some BMPs (Hemmati-Brivanlou et al., 1994; Yamashita et al., 1995), on the right side of the forming node
at stage 3. In 5 of 20 cases, Shh, which is normally repressed
on the right side of Hensen’s node (Fig. 3A), was symmetrically expressed on both sides of the node following follistatin application (Fig. 3B). Preliminary experiments indicate that symmetrical nodal expression can also result from
follistatin treatment (data not shown). Neither Shh nor
nodal was ever expressed bilaterally following control bead
implants (n Å 23).
To identify a candidate for the endogenous activin-like
signal, we examined the expression of various activin and
related BMP genes at stage 4 (including BMP-2, 4, 6, and 7
and Activin bA and bB), all of which except one either were
TABLE 3
Right-sided (wt) hearts
Left-sided (reversed) hearts
Bilaterally symmetric hearts
No. of embryos examined
Statistics
Control cells
on right side
Nodal cells
on left side
Nodal cells
on right side
81%
10%
9%
31
82%
0%
18%
22
x2 Å 5.3, P Å 0.07
38%
19%
43%
21
x2 Å 23.9, P Å 0.005
Note. Pellets of CEFs infected with the nodal virus were implanted into the right side of Hensen’s node of stage 6 embryos in New
culture. Embryos were allowed to develop and the situs of the heart was scored.
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$403
08-12-97 06:09:05
dbal
62
Levin et al.
FIG. 2. Nodal determines heart asymmetry. Ventral views: when
nodal is misexpressed on the right side of the node (n Å 21), 38%
of the resulting hearts are wt (A), while 19% show left-sided looping
(B) and 43% are symmetrical (C). However, when cells infected
with a control virus (alkaline phosphatase) are implanted on the
right side of the node (n Å 31), 81% of the resulting hearts are wt,
10% exhibit left-sided looping, and 9% are symmetrical. This effect
is significant to P Å 0.005, x2 Å 23.9. When a nodal-expressing
pellet is implanted on the left side of the node (n Å 22), 82% of
the hearts are wt, 0% are left-sided, and 18% are symmetrical.
These phenotypes are not statistically significantly different from
control implants (P Å 0.07, x2 Å 5.3). Arrows indicate looping of
the heart tube.
not detectable at stage 4 or were expressed symmetrically
(BMP-6 was not detectable, while BMPs 2, 4, and 7 were
expressed in the posterior third of the primitive streak, neural folds, and lateral mesoderm, respectively, data not
shown). Chick Activin bB was seen in whole-mount in situ
hybridization to be specifically expressed on the right side
of Hensen’s node from stage 3 to stage 5/. Unfortunately
the Activin bB probe gives a high, uniform, nonspecific
background at all stages examined. Nonetheless, the rightsided expression can be clearly seen above background,
most clearly at stage 4/ (Fig. 3C). To verify that this signal
was real, and was present prior to the asymmetry in Shh
expression, we sectioned several entire stage 3 embryos in
a plane perpendicular to the primitive streak and hybridized
all sections with an Activin bB probe. Hybridization was
exclusively detected on the right side in sections through
the anteriormost tip of the primitive streak (Fig. 3D). Thus,
the Activin bB gene is specifically expressed on the right
side of the node at the correct stage to influence Shh expression and is therefore a candidate for the endogenous activinlike signal in the LR cascade.
Asymmetry Does Not Appear to Arise from a
PrePatterned Blastoderm
Activin bB is currently the earliest molecular marker of
LR asymmetry in the chick. This leaves unexplored the
factors responsible for initiating LR asymmetries in gene
expression in the chick embryo. At least two models for
this have been proposed. One theory (Brown and Wolpert,
1990) suggests that LR asymmetry is determined within
each cell (perhaps by a chiral molecule which is oriented
with respect to the AP and DV axes, which are well established by stage 2 in the chick). An alternative hypothesis is
that LR information arises from the maternal localization
of a determinant in a LR asymmetric manner within the
blastoderm (Wilhelmi, 1921). The analysis of spontaneous
twin chick embryos gives a possible indication of which of
these two hypotheses about the initial origin of LR asymmetry is more likely to be correct. The cell-autonomous theory
predicts that in twin embryos that are oriented in a headto-head fashion (1807), each twin will be correctly patterned
along the LR axis because each node will contain cells
which know left from right with respect to their own AP
orientation (Fig. 4A). The prepattern theory, however, predicts that the twins will be mirror images of each other
since the LR factors localized in the blastoderm will determine left and right sides regardless of the AP orientation
of the embedded streaks, resulting in one correct and one
reversed embryo (Fig. 4B). Examination of four such embryos probed with Shh (Fig. 4C) and nodal (data not shown)
as markers of laterality, as well as six such sets of twins
assayed by the morphology of the node at stage 5 (Fig. 4D;
Cooke, 1995) and two sets of twins examined for embryo
turning at stage 22 (Fig. 4E), shows that, in every such case,
each embryo is correctly patterned with respect to its own
orientation. This suggests that LR information is initiated
with respect to the AP and DV axes of each embryo and is
due neither to a maternally derived prepattern within the
blastodisc nor to a zygotic decision prior to the establishment of the AP and DV axes.
DISCUSSION
The Shh Pathway and Heterotaxia
Shh has been previously shown to randomize heart situs
(Levin et al., 1995). We now find that when ectopic (rightsided) Shh protein is applied to the node in ovo, reversals of
the situs of the heart, gut, and embryonic rotation are observed. Thus, Shh does not lie upstream of a heart-specific
pathway, but rather provides a left–right reference by which
multiple organs assess their laterality during morphogenesis.
In normal development, the morphogenesis of different
organs is a tightly coordinated process. For the asymmetric
organs to have an invariant orientation relative to each
other, their primordia must respond to a common LR asymmetric set of cues. This is verified by the existence of situs
inversus mutants (e.g., the inv mouse; Yokoyama et al.,
1993) where all of the internal organs are in an absolute
reverse orientation but maintain the same relative configuration. On the other hand, ultimately each organ forms independently, and other, presumably downstream, mutations result in a randomization of asymmetric placement
of each organ. This is seen in human heterotaxia syndromes
(e.g., Afzelius, 1976).
In principle, asymmetric signals could affect all of the
organ systems by initially acting on one, such as the heart
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$403
08-12-97 06:09:05
dbal
63
Left/Right Patterning Signals
FIG. 3. Endogenous asymmetric activin-like activity. Control beads (soaked in PBS) or beads carrying follistatin protein were implanted
into the right side of Hensen’s node at stage 3, and the embryos were harvested at stage 5 and processed for in situ hybridization with
the Shh probe. Control beads never caused symmetrical expression patterns of Shh (n Å 23, A). In contrast, when a bead loaded with
follistatin protein was implanted in the same manner, symmetrical expression of Shh is observed (5/20 cases, B). This result is significant
to P Å 0.02. The beads move anteriorly from their original implant location because of the cell migration which occurs during incubation
(see B, green arrowhead). The bead in A is not visible because it became dislodged during in situ hybridization processing. Stage 4/ embryos
in whole mount (C) and cryosections at the level of the forming node of stage 3 embryos (D), were processed for in situ hybridization
with the Activin bB probe. Signal was detected in the right side of Hensen’s node (black arrow), consistent with its proposed role in
repressing Shh there. Black arrows indicate endogenous expression domain. Gray arrows indicate ectopic domain. L and R, left and right
sides of the primitive streak, respectively.
FIG. 4. LR asymmetry is apparently streak autonomous. The maternal prepattern theory suggests that such twins should result in mirror
image embryos because the blastodisc is divided into left and right domains (shown here as yellow and white) containing distinct positional
information, such as an asymetrically positioned factor which influences subsequent LR decisions (B). The chiral molecule theory predicts
that embryos which are arranged head to head should have correct LR orientation with respect to their own axes (A). When such twins
are hybridized to a Shh probe, it is seen that each embryo is correctly patterned with respect to itself (C). Red arrows point to wt leftsided Shh expression. The morphological asymmetry in the node, shown in section through a wt embryo (D, black arrowhead points to
node asymmetry), as well as the direction of embryonic turning (shown in E, black arrowheads point to anterior) is likewise correct in
each twin, with respect to its own AP and DV axes.
primordia, which could then mechanically influence the
others (Waddington, 1937). In such a case, heterotaxia could
be explained by a decoupling of different organs from this
influence. Alternatively, each organ primordia could directly respond to the asymmetric signals, and heterotaxia
would be a consequence of the failure to interpret these
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$403
08-12-97 06:09:05
dbal
64
Levin et al.
cues (Brown and Wolpert, 1990). Our experiments support
the latter scenario (consistent with the findings of Fujinaga
and Baden, 1991b) since ectopic application of Shh results
in independent alteration of situs of each of the properties
we assayed.
In these experiments we placed Shh protein on the right
side of Hensen’s node resulting in bilateral Shh signaling
and hence bilateral nodal expression (Levin et al., 1995).
Bilateral nodal expression has also been correlated with independent segregation of heart and gut orientation in the
frog, following ectopic Vg1 expression (Hyatt et al., 1996).
It remains to be determined which organs respond to this
nodal signal directly and which if any organs get their LR
information from genes upstream of (Shh), downstream of,
or parallel to nodal (for example, lefty; Meno et al., 1996).
The fact that heterotaxia was observed in these experiments
instead of left isomerism (which might have been expected
to result from the double-sided expression of Shh, a normally left-sided gene), suggests that the Shh pathway is
involved in the biasing of random asymmetry, not in its
generation (Brown and Wolpert, 1990).
Nodal Is a Causal Determinant of Heart Situs
Randomization of heart situs caused by ectopic Shh or
activin has been correlated with double-sided and absent
nodal expression respectively (Levin et al., 1995). Similarly,
recent studies showed that expression of nodal is altered
in two mouse mutations (iv, Lowe et al., 1996; and inv,
Collignon et al., 1996) which also display laterality defects.
This placed the mutations upstream of nodal, but it did not
distinguish between nodal’s being a causal factor in heart
situs determination and its expression being a parallel effect
of earlier parts of the asymmetric gene cascade. Here, we
show that ectopic nodal expression itself results in inverted
and double-sided hearts. This strongly suggests that nodal
endogenously controls the laterality of cardiac looping.
In these experiments, a significant percentage of embryos
display right-isomerized symmetric hearts following bilateral nodal expression achieved by implanting nodal-expressing cells. This is a striking contrast to the phenotypes
observed following bilateral nodal expression generated by
implanting Shh-expressing cells, in which heart laterality
was randomized but no symmetric hearts were observed
(Levin et al., 1995). One possible explanation for this difference could be that nodal is but one component of the signals
downstream of Shh necessary to fully specify heart situs.
For example, lefty, another TGF-b family member, has been
identified in mice (Meno et al., 1996), has an expression
pattern similar to that of nodal, and may work in concert
with it. An alternative interpretation is that ectopic and
endogenous nodal domains, when induced by Shh, are only
transiently expressed in the cardiac-forming region; in contrast, nodal-expressing cell implants come to lie next to the
heart tube and provide a constant source of signal adjacent
to the forming heart, which can in some instances disrupt
its morphogenesis. Consistent with this explanation, we
also observe an increase in bilaterally symmetric hearts
when nodal cell implants are placed on the left side, where
nodal is normally expressed.
Nodal, a member of the TGF-b superfamily, encodes a
secreted factor (Zhou et al., 1993, and data not shown).
Thus, it is plausible that it directly affects heart looping by
providing an asymmetric signal to only one of the heart
primordia. This could then affect heart morphogenesis by
affecting the migration (Manasek, 1981), proliferation
(Stalsberg, 1969), or cytoskeletal organization (Itasaki et al.,
1989, 1991) of cells descended from the left-side cardiac
precursors.
An apparent paradox arises: Hoyle et al. (1992) report a
difference (in the ability to bias the heart tube) between the
left and right precardiac mesoderm as early as stage 5–6,
while nodal is not expressed until somewhat later. The
resolution of this issue is likely to be that cells become
committed to express nodal earlier, at the time when Shh is
expressed asymmetrically (stage 4/). Consistent with this,
when cells expressing Shh are ectopically implanted on the
right side adjacent to Hensen’s node, the cell pellet migrates
anteriorly prior to nodal expression. Nodal is then induced
next to the location where the Shh cells were originally
implanted, demonstrating a commitment to express nodal
prior to its actual expression. This commitment is probably
what Hoyle et al. observed, assayed by them as the ability
to bias the heart tube when transplanted.
An Endogenous Activin-like Signal Is Upstream of
Asymmetric Shh Expression
Shh is asymmetrically expressed in the early chick embryo and is capable of inducing asymmetric nodal expression. The identification of this pathway of genes which are
asymmetrically expressed leads to the question of further
upstream factors: what is responsible for the asymmetry in
the expression of Shh? Several lines of evidence suggest that
an activin-related signal may play a critical role. First, previous studies demonstrated that exogenously applied activin
is sufficient to repress Shh on the left side of the node and
thereby also prevent nodal induction (Levin et al., 1995).
In other reported experiments, exogenous activin had somewhat different consequences for Shh expression, resulting in
some cases of reversed Shh expression as well as bilaterally
expressed Shh (Isaac et al., 1997). The reason for this difference is unclear, but may reflect differences in timing or
placement of the activin beads. In any case, both sets of
experiments indicate that exogenous activin can act upstream of Shh. Here we provide evidence that this pharmacological effect likely reflects the action of a related endogenous signal normally present in the right side of the node,
since ectopic follistatin, an inhibitor of activin and related
factors, leads to symmetrical Shh expression. An excellent
candidate for the endogenous Shh-repressing activity is Activin bB. Activin bB is expressed in the right side of the
node just before and during the expression of cAct-RIIa
there and, most importantly, before the disappearance of
Shh expression from the right side. However, it should be
noted that follistatin may also interfere with signaling by
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$403
08-12-97 06:09:05
dbal
65
Left/Right Patterning Signals
related molecules such as BMP-7 (Yamashita et al., 1995,
Wilson and Hemmati-Brivanlou, 1995) and hence the endogenous activin-like activity upstream of Shh may in fact
be a related molecule. We examined early chick embryos for
the expression of a number of related signaling molecules
(including BMP-2, 4, 6, and 7) and none were expressed in
the node, or in a manner that would indicate a role in LR
patterning. However, it remains possible that another member of this family exists which is asymmetrically expressed
on the right side of Hensen’s node at the same time as
Activin bB. In any case, asymmetric expression of Activin
bB is the earliest currently known marker of left–right
asymmetry in the chick, expressed asymmetrically long before major organ LR asymmetry. The events further upstream that initiate LR asymmetry and lead to right-sided
Activin bB expression remain enigmatic.
Vg1 has been reported to be a possible candidate for a
signal upstream of activin (Hyatt et al., 1996). BVg1 misexpressed on the right side of Xenopus embryos, or expression
of a dominant-negative (truncated) activin receptor on the
left side, produces situs defects in the resulting embryos. It
should be noted however, that the truncated activin receptor interacts with other ligands (Kessler and Melton, 1995;
Hemmati-Brivanlou et al., 1995); likewise, injected BVg1
may cross-react with receptors for other TGF-b family
members. Thus it is possible that these results reflect manipulation of an activin signal, not endogenous Vg1 signaling. It is unlikely that the experiments presented here reflect the activity of an endogenous chick Vg1 homologue,
as Vg1 signaling is not inhibited by follistatin (Kessler and
Melton, 1995).
Activin and its receptors have not been reported to be
asymmetric in any species other than chick. Moreover, several mice have been generated which carry null mutations
for Activin-bB and Activin receptor IIa (Matzuk et al.,
1995), and these mice appear to have no phenotype associated with LR patterning. However, null mutations in Activin receptor IIb do result in laterality defects (En Li, personal communication), suggesting that this part of the pathway may indeed also be conserved in mammals. The target
of asymmetrical activin signaling in chicks is Shh, but Shh
does not appear to be asymmetric in the mouse node (Collignon et al., 1996), and mice carrying a homozygous deletion
of Shh have no laterality defects (Chiang et al., 1996). These
differences between chicks and mice may be due to different
homologues playing the respective roles in LR signaling, or,
perhaps the early steps involving activin and Shh are specific to avian species.
There Is Not an Irreversible LR Determination
Prior to the Induction of the Primary Axis
The series of experiments described here is concerned
with the investigation of the middle part of LR patterning:
the cascade of differential gene expression which lies between initial LR asymmetry determination (the cause of
the restriction of the very first gene to be asymmetrically
expressed in any given embryo) and the final asymmetric
morphogenesis of organs. The two most commonly discussed mechanisms of initial LR determination, a blastodermal prepattern of maternal origin and a chiral molecule
within each cell, make opposite predictions for the laterality of 1807 head-to-head twins. We have examined 12
cases of such spontaneously occurring twins. Determination of situs by means of molecular markers of laterality,
embryonic turning, and morphological asymmetry at
Hensen’s node revealed that each twin was correctly patterned with respect to itself, not to a hypothesized blastodermal prepattern.
Although it is not clear with spontaneously arising twins
exactly when the secondary streak was formed, in all cases
examined the 1807 twins appeared to be of identical stage
and size, indicating that they arose at nearly the same time.
In principle such spontaneous twins could have arisen
within a common blastodisc or in two separate blastodiscs
which subsequently fused. However, experimental protocols which induce 1807 twins from within the same blastodisc give similar results to those we obtained, at least as
assayed morphologically. For example, mechanical transection of a blastoderm can lead to formation of such head-tohead twins, each of which has correct situs (Lepori, 1967).
Likewise, induction of ectopic streaks with activin and Wnt
proteins (Cooke et al., 1994) can produce such twins, which
we found in preliminary experiments to give similar results
to those described here. We chose not to pursue the experimentally induced twins because in that paradigm, the
streak-inducing factors could themselves influence situs,
making interpretation difficult.
Our examination of 1807 twins suggests that their respective situs either is determined without regard to information in the blastodisc or is dominant to any underlying bias
present as a blastodisc prepattern. This streak-autonomous
model is consistent with the theory of Brown and Wolpert
(1990) who propose that cells contain a tethered chiral molecule whose directed differential activity serves to produce
LR asymmetries. The rarity of experimental material made
it impossible to obtain sufficient numbers to confidently
test a variety of other angular orientations in addition to
1807. This would have been useful to rule out more exotic
prepattern geometries than the one set out in Fig. 4; however, the results of Lepori (1967), who produced duck twins
in various angular orientations and observed very few cases
of situs inversus, are consistent with our own. The one
other relatively common class of spontaneous twins represents those arising in parallel or nearly parallel orientations.
Under such circumstances the nodes and streaks of the two
embryos are in close juxtaposition throughout their development and in those cases the asymmetric signals appear
to be able to cross between embryos (Levin et al., 1996). In
the 1807 twins the two primitive streaks form at opposite
ends of the blastodisc. The nodes never become closely juxtaposed because even at full extension the primitive streak
does not reach the anterior limit of what will be the embryo.
We do not see evidence for cross-signaling in the 1807 twins
as downstream asymmetric markers and morphology are
normal in each twin.
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$404
08-12-97 06:09:05
dbal
66
Levin et al.
ACKNOWLEDGMENTS
We thank Susan Smith and Devon Smith for advice on in ovo
surgery, Tiffany Heanue, Christine Hartmann, and other members
of the Tabin and Cepko labs for providing spontaneous twin chick
embryos, and Ketan Patel for providing the Activin bB clone. Follistatin protein was obtained from Dr. Philip Smith at the National Hormone and Pituitary Program. Sonic hedgehog protein
was provided by Ontogeny, Inc. This work was funded by a grant
from the NIH.
REFERENCES
Afzelius, B. A. (1976). A human syndrome caused by immotile cilia.
Science 193, 317–319.
Brown, N., and Wolpert, L. (1990). The development of handedness
in left/right asymmetry. Development 109, 1–9.
Chiang, C., Ying, L. T. T., Lee, E., Young, K. E., Corden, J. L., Westphal, H., and Beachy, P. A. (1996). Cyclopia and defective axial
patterning in mice lacking sonic hedgehog gene function. Nature
383, 407–413.
Collignon, J., Varlet, I., and Robertson, E. J. (1996). Relationship
between asymmetric nodal expression and the direction of embryonic turning. Nature 381, 155–158.
Cooke, J. (1995). Vertebrate embryo handedness. Nature 374, 681.
Cooke, J., Takada, S., and McMahon, A. (1994). Experimental control of axial pattern in the chick blastoderm by local expression
of Wnt and activin: The role of HNK-1 positive cells. Dev. Biol.
164, 513–527.
Danos, M. C., and Yost, H. J. (1995). Linkage of cardiac left-right
asymmetry and dorsal–anterior development in Xenopus. Development 121, 1467–1474.
Freeman, G., and Lundelius, J. W. (1982). The developmental genetics of dextrality and sinistrality in the gastropod Lymnaea peregra. Wilhelm Roux Arch. 191, 69–83.
Fujinaga, M., and Baden, J. M. (1991a). Evidence for an adrenergic
mechanism in the control of body asymmetry. Dev. Biol. 143,
203–205.
Fujinaga, M., and Baden, J. M. (1991b). Critical period of rat development when sidedness of body asymmetry is determined. Teratology 44, 453–463.
Halpern, M. E., Ho, R. K., Walker, C., and Kimmel, C. B. (1993).
Induction of muscle pioneers and floor plate is distinguished by
the zebrafish no tail mutation. Cell 75, 99–111.
Hamburger, V., and Hamilton, H. L. (1951). A series of normal
stages in the development of the chick embryo. J. Morphol. 88,
49–92.
Hemmati-Brivanlou, A., Kelly, O. G., and Melton, D. A. (1994). Follistatin, an antagonist of activin, is expressed in the Spemann
organizer and displays direct neuralizing activity. Cell 77, 283–
295.
Hemmati-Brivanlou, A., and Thomsen, G. H. (1995). Ventral mesodermal patterning in Xenopus embryos: Expression patterns and
activities of BMP-2 and BMP-4. Dev. Genet. 17, 78–89.
Hoyle, C., Brown, N. A., and Wolpert, L. (1992). Development of
left/right handedness in the chick heart. Development 115,
1071–1078.
Hughes, S. H., Greenhouse, J. J., Petropoulos, C. J., and Sutrave, P.
(1987). Adapter plasmids simplify the insertion of foreign DNA
into helper-independent retroviral vectors. J. Virol. 61, 3004–
3012.
Hummel, K. P., and Chapman, D. B. (1959). Visceral inversion and
associated anomalies in the mouse. J. Hered. 50, 9–23.
Hunt, P., and Krumlauf, R. (1992). Hox codes and positional specification in vertebrate embryonic axes. Annu. Rev. Cell Biol. 8,
227–256.
Hyatt, B. A., Lohr, J. L., and Yost, H. J. (1996). Initiation of vertebrate left–right axis formation by maternal Vg1. Nature 384, 62–
65.
Isaac, A., Sargent, M. G., and Cooke, J. (1997). Control of vertebrate
left-right asymmetry by a snail-related zinc finger gene. Science
275, 1301–1304.
Itasaki, N., Nakamura, H., and Yasuda, M. (1989). Changes in the
arrangement of actin bundles during heart looping in the chick
embryo. Anat. Embryol. 180, 413–420.
Itasaki, N., Nakamura, H., Sumida, H., and Yasuda, M. (1991). Actin bundles on the right side in the caudal part of the heart tube
play a role in dextro-looping in the embryonic chick heart. Anat.
Embryol. 183, 29–39.
Ivemark, B. (1955). Implications of agenesis of the spleen on the
pathogenesis of conotruncal anomalies in childhood. Acta Paediatr. Scand. 44, 590.
Jowett, T., and Lettice, L. (1994). Whole-mount in situ hybridization of zebrafish embryos using a mixture of digoxigenin- and
fluorescein-labeled probes. Trends Genet. 10, 73–74.
Kessler, D. S., and Melton, D. A. (1995). Induction of dorsal mesoderm by soluble, mature Vg1 protein. Development 121, 2155–
2164.
Klessinger, S., and Christ, B. (1996). Axial structures control laterality in the distribution pattern of endothelial cells. Anat.
Embryol. 193, 319–330.
Layton, W. M., Binder, M., Kurnit, D. M., Hanzlik, A. J., Van Keuren, M., and Biddle, F. G. (1993). Expression of the IV (reversed
and/or heterotaxic) phenotype in SWV mice. Teratology 47, 595–
602.
Lepori, N. G. (1969). Sur la genèse des structures asymmétriques
chez l’embryon des oiseaux. Monit. Zool. Ital. 3, 33–53.
Levin, M., Johnson, R., Stern, C., Kuehn, M., and Tabin, C. (1995).
A molecular pathway determining left–right asymmetry in chick
embryogenesis. Cell 82, 1–20.
Levin, M., Roberts, D. J., Holmes, L. B., and Tabin, C. (1996). Laterality defects in conjoined twins. Nature 384, 321.
Lowe, L., Supp, D.-M., Sampath, K., Yokoyama, T., Wright, C. V. E.,
Potter, S. S., Overbeek, P., and Kuehn, M. R. (1996). Conserved
left–right asymmetry of nodal expression and alterations in murine situs inversus. Nature 381, 158–161.
Manasek, F. J. (1981). Determinants of heart shape in early embryos. Fed. Proc. 40, 2011–2016.
Matzuk, M. M., Kumar, T. R., Vassalli, A., Bickenbach, J. R., Roop,
D. R., Jaenisch, R., and Bradley, A. (1995). Functional analysis of
activins during mammalian development. Nature 374, 354–356.
McCain, E. R., and McClay, D. R. (1994). The establishment of bilateral asymmetry in sea urchin embryos. Development 120,
395–404.
Meno, C., Saijoh, Y., Fujii, H., Ikeda, M., Yokoyama, T., Yokoyama,
M., Toyoda, Y., and Hamada, H. (1996). Left–right asymmetric
expression of the TGFb-family member lefty in mouse embryos.
Nature 381, 151–154.
Neville, A. C. (1976). ‘‘Animal Asymmetry.’’ Arnold, London.
New, D. A. T. (1955). A new technique for the cultivation of the
chick embryo in vitro. J. Embryol. Exp. Morphol. 3, 326–331.
Riddle, R. D., Johnson, R. L., Laufer, E., and Tabin, C. (1993). Sonic
hedgehog mediates the polarizing activity of the ZPA. Cell 75,
1401–1416.
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$405
08-12-97 06:09:05
dbal
67
Left/Right Patterning Signals
Sampath, K., Cheng, A. M. S., Frisch, A., and Wright, C. V. E. (1997).
Functional differences among Xenopus nodal-related genes in
left-right axis determination. Development, in press.
Schultheiss, T., Xydas, S., and Lassar, A. B. (1995). Induction of
avian cardiac myogenesis by anterior endoderm. Development
121, 4203–4214.
Stalsberg, H. (1969). The origin of heart asymmetry: Right and left
contributions to the early chick embryo heart. Dev. Biol. 19,
109–127.
Stalsberg, H., and DeHaan, R. L. (1969). The precardiac areas and
formation of the tubular heart in the chick embryo. Dev. Biol.
19, 128–159.
Talbot, W. S., Trevarrow, B., Halpern, M. E., Melby, A. E., Farr, G.,
Postlethwait, J. H., Jowett, T., Kimmel, C. B., and Kimelman,
D. (1995). A homeobox gene essential for zebrafish notochord
development. Nature 378, 150–157.
Tuinstra, E. J., De Jong, G., and Scharloo, W. (1990). Lack of response to family selection for directional asymmetry in Drosophila melanogaster: Left and right are not distinguished in development. Proc. R. Soc. London Series B 241, 146–152.
Waddington, C. H. (1937). The dependence of head curvature on
the development of the heart in the hick embryo. J. Exp. Biol.
14, 229–231.
Wilhelmi, H. (1921). Experimentelle untersuchungen uber situs inversus viscerum. Arch. Entwicklungsmech. Org. 48, 517–532.
Wilson, P. A., and Hemmati-Brivanlou, A. (1995). Induction of epidermis and inhibition of neural fate by Bmp-4. Nature 376, 331–
333.
Winer-Muram, H. T. (1995). Adult presentation of heterotaxic
syndromes and related complexes. J. Thoracic Imaging 10, 43 –
57.
Yamashita, H., Tendijke, P., Huylebroeck, Sampath, T., Andries,
M., Smith, J. C., Heldin, C-H., and Miyazono, K. (1995). Osteogenis protein-1 binds to activin type II receptors and induces
certain activin-like effects. J. Cell Biol. 130, 217–226.
Yokoyama, T., Copeland, N. G., Jenkins, N. A., Montgomery,
C. A., Elder, F. F., and Overbeek, P. A. (1993). Reversal of left–
right asymmetry: A situs inversus mutation. Science 260, 679–
682.
Zhou, X., Sasaki, H., Lowe, L., Hogan, B. L., and Kuehn, M. R.
(1993). Nodal is a novel TGF-beta-like gene expressed in the
mouse node during gastrulation. Nature 361, 543–547.
Received for publication April 10, 1997
Accepted June 10, 1997
Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.
AID
DB 8662
/
6x2a$$$405
08-12-97 06:09:05
dbal